Adrian Taylor

Current state of play: * `bindgen` doesn't correctly identify that some types are templated, per https://github.com/rust-lang/rust-bindgen/issues/1924. I'm working on this in the background. For such types, we don't really stand...

Current state of play here with respect to Chromium: * https://chromium-review.googlesource.com/c/chromium/src/+/2784885 made that we no longer suffer this problem with `base::StringPiece` (thanks Jan!) * But unfortunately `base::Optional` still seems to...

I posted a status update to https://github.com/rust-lang/rust-bindgen/issues/1924

I'm removing the "bug" label here since we now gracefully avoid mis-generations in such cases. As such this now becomes a C++ feature for which we need to add support.

I'd accept a PR for clone, yep. I am not so sure about copy. I think it's probably best to require explicit cloning.

At the moment, we implement a copy constructor for any type which _can_ be copied by implementing moveit's [`CopyNew` trait](https://docs.rs/moveit/latest/moveit/trait.CopyNew.html). This applies whether the type has an explicit copy constructor...

I see at least four different bugs when trying to build this (at least on OS X). At some point in the future I'll aim to run each one through...

Working through #1069 should allow us eventually to land this.

Thanks for this. I haven't managed to get the right versions of libclang to reproduce it yet. I am pretty sure this would be an upstream issue in `bindgen` but...

We already support `rust::Str` and `rust::String` but still need a fair bit of work to support `rust::Box`, `rust::Slice` and opaque Rust types.